Alternative splicing is a widespread mechanism for post-transcriptional control of gene expression, and accounts for a large fraction of proteomic diversity. This process is also frequently misregulated in cancer, and appears to contribute to various changes associated with transformation. This project explores the molecular mechanisms through which the members of two conserved families of RNA-binding proteins regulate alternative splicing, and the range of pre-mRNA targets each of them controls in the context of transformation. Overexpression of these factors can promote tumorigenesis, and apparently bypass the requirement for oncogene cooperation by controlling the expression of specific isoforms of several key members of oncogene and tumor-suppressor networks. The prevalence of this mechanism for tumor development in different types of human cancer will be studied, and its specific features will be dissected and compared in different contexts by genetic manipulation of the splicing factors in cell culture and in models of B-cell lymphoma, hepatocellular carcinoma,and breast carcinoma. Distinctive features ofpre- mRNA targets recognized by the splicing machinery will also be studied, both to understand the mechanism and specificity of this process and to facilitate the classification of mutations in cancer- susceptibility genes that result in defective mRNA and protein. By exploring a new pathway for tumor development,this study may define markers that facilitate early detection of genetic lesions leading to cancer, and it may also potentially uncover novel targets for cancer therapy. Improved prediction of mutations that cause defective RNA splicing will inform treatment decisions for individuals with genetic predisposition to cancer.